Symbolic Regression Data-Model-Informed Eco-Control for Maritime Carbon Emissions Reduction

This paper proposes eco-control optimization with consideration of the physics model informed by data-model to emphasize the maritime carbon emissions reduction problem. More specifically, our work centers on developing and utilizing physics and symbolic regression data-model that capture the dynamics of ship engines and vessels. These models are then integrated with the optimization algorithm to plan the optimal trajectories. The primary objective is to enhance the environmental sustainability of maritime operations by significantly decreasing carbon emissions. To this end, we explore various aspects of ship dynamics, including the interactions between the hull, propeller, and engine, especially under the influence of regular wave patterns in acceleration mode. Symbolic regression is used to model fuel consumption parameters. Our approach leverages the power of optimized control, taking into account the continuous variability in external conditions and the need for adaptive response mechanisms. The results demonstrate that this method is critical in achieving greener maritime transportation without compromising on operational efficiency. This study not only contributes to the theoretical advancements in eco-friendly ship control but also provides a practical framework for implementing these strategies in real-world maritime operations.